Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Works up to a certain number of characters and then freezes #1

Open
rocketraman opened this issue Apr 27, 2016 · 3 comments
Open

Works up to a certain number of characters and then freezes #1

rocketraman opened this issue Apr 27, 2016 · 3 comments

Comments

@rocketraman
Copy link

I'm trying to benchmark an editor and see that typometer seems to benchmark consistently up to a certain number of characters and then it freezes.

At the normal zoom level I was able to benchmark 15 characters, and with a higher zoom level I got up to 68 characters. But it consistently freezes on the same count.

I've followed the recommendations in the "common problems" section.

@pavelfatin
Copy link
Owner

The symbol detection algorithm implies that distance between the consecutive dots is uniform. If that is not the case, the algorithm might look for for a newly typed dot in a wrong place of the screen, which finally results in a time out. You can manually type a long series of dots in the editor, make a screenshot and check the distances in image editor.

Another possibility is that the image recognition algorithm detects screen metrics incorrectly. In this case, you can either try to create a new test with the specific images (examples) or just send me an image of a long sequence of dots so I can try to figure out the underlying cause.

@hmage
Copy link

hmage commented Sep 19, 2017

I think there's a way to do it -- fill the entire line with dots instead of just five dots. No need to extrapolate positions for rest of the line, since you know where all of them should be showing up.

@frarees
Copy link

frarees commented Oct 8, 2020

Increasing the pattern length as @hmage suggested works, since the average distance gets accurate enough. Filling the whole line would be the safest. frarees@ea57321 did it for me, but I think it'd be nice to expose this values to the tool itself for proper fine-tuning.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants